Week 4 Recap: Night vs. Day. Cheddar vs. Bleu. As expected the level of play was much higher at Central Virginia than at Central Maryland. Some of this stemmed from the quality of robot but some of it was a result of Central Virginia teams pulling together the materials for the new peg design and Central Maryland making the decision to hold off.
Teams 384 and 1885 were involved in 3 of the 4 qualification matches that activated the 4th rotor and teamed up to activate another 4 more in the playoffs on their way to win number 2 of the season for each team. Their win in the finals was not an easy one as the #2 alliance led by 1262 took the round to 4 matches using two robots on defense.
At Central Maryland 5115 picked up 1389 and 2534 as the #5 captain and were the alliance of consistency averaging 297 points through the playoff rounds and 305+ to upset the #2 alliance led by rookie 6239. 6239 put up an impressive showing ranking #2 after having missed the playoffs just the week before at Northern Maryland
Interesting CHSy Fact: CHSyness was 78.5% accurate at predicting the 24 teams to make the playoff rounds.
CHS Platter - Week Six - Virginia is for Lovers
CHS Splatter - What happens when someone in CHS eats too much cheese that is full of Jelly.
Extra Sharp Prediction: The CHSy Champ finals will see 8 rotors, 6 climbs, and will be decided by fuel.
Chesapeake District Championship
Number of Registered Teams: 58
TBA page: https://www.thebluealliance.com/event/2017chcmp
Livestream: https://livestream.com/LeeHartman/DistrictChampionship
CHSyness: 65(+20)
Ranking Predictions:
The mice behind the CHS Platter have been fed more Espresso BellaVitano than usual during their two week break from interesting, hawkish predictions. In that time they’ve come up with a few hundred lines of prediction code that simulate actual match scores in 100,000 random schedules based upon “capped” component OPR’s. “Capped” OPR takes into account that as a collaborative task is more likely to be achieved, the “contributed value” of each individual approaches the total value of the task divided by the number of teams who contributed. For example, whether or not we believe it conveys the wrong message to our Youth, Rotor 1 is a freebie. So the component OPR’s for each team for rotor 1 is exactly 13.3 (repeating). The Platter applied a cap of 13.3 (non-repeating) to the 2nd and 3rd rotor.
The component OPR’s for autonomous moves & teleop climbs are capped to an individual value of 5 & 50 points, respectively. Fuel is un-capped.
In order to attribute great alliances coming together for the 4th rotors, Rotor 2 & Rotor 3’s “extra” OPR roll forward into the next rotor. For example, say 384 & 836 both have Rotor 2 OPR’s of 16. The algorithm takes the difference between their OPR values and the “capped” value (16 - 13.3) and adds that to their 3rd rotor OPR. The same happens for the 3rd rotor, rolling the “extra” OPR into the 4th rotor. Rotor’s 2 & 3 are predicted to be activated if a random number is less than the maximum probability of activation on the alliance. The maximum probability is (13.3 * 3) / 40, or 99.75% - this helps to account for very random unlucky circumstances. Rotor 4 is predicted to be activated if a random number is less than the average probability of activation on the alliance. We average the 4th-rotor probabilities to account for the likely defense that can force a single great robot to miss the 4th rotor. This methodology correctly approximated the rate of Israel’s & North Carolina’s 4-rotor qual matches in a field of robots where nearly all of them had never executed a 4-rotor match in qualifications before district champs. The overall rankings had a margin of error of about 10% (4-6 teams) at 80% confidence.
So The Platter has some component OPR’s and some predicted rankings. While the OPRs are easily calculated, the rankings should be taken with a grain of salt. Here’s why:
There is a 3-way statistical tie in the CHS platter predictions for 1st seed, and a 6-way statistical tie for 5th seed. On top of that, any team predicted to be in the top 10 has a good chance of being 1st seed with an average schedule and any improvement to their prior performance. Any team predicted to be rank 29 or lower has a fairly good chance to be a captain with a lucky schedule. Yet with an unlucky schedule, even the best teams could find themselves ranked in the lower half of all teams due to lack of kPa or consistent 4-rotor bonus.
Expect the autonomous rotor to be the difference-maker in 17% of qual matches. Expect 1-2 fuel points to be a tiebreaker in about 4% of qual matches, but expect what would be fuel tiebreakers to be overcome by a 4th rotor in a further 8% of matches. And that prediction is based upon the statistics of most teams that have yet to experience the peg upgrades that enabled the reign of 4 rotors at Central Virginia. Barring a 10% improvement in the district’s top fuel scorer, we don’t foresee any kPa bonus matches. However, we do foresee at least 13% of qual matches earning the Rotor RP bonus.
How to Win CHS Champs:
Expect to need scores Gouda than 285 or more in order to seed anywhere in the top 8. That means every team moves their cheese wheels in autonomous, fills up the cheese fountain to at least the 3rd level, and then climbs every match. Look for the X-Factor in your robot and your qualifications partners’ robots, as that is what ultimately made the Finals in Glen Allen more interesting to watch than the cheese hats. Eschew defense in matches where your alliance is capable of the 4th rotor, since prior events have proven that defense against 3 rotors is like Swiss: tempting, tasty and ultimately holey. Finally, given the impact of a moldy schedule on ranking and stats, **scouting **will be one of the most significant things the winning alliances bring to the table. Mere OPR statistics and event rankings will not tell you which team is the best fit for your alliance, as is demonstrated by the attachments.
The Favorites:
1885 (88) - With two event wins this season from a robot that could execute gear cycles under heavy defense, ILITE looks to make up for missing last year’s DCMP tournament. Great scouting, an active gear intake and a rock’em sock’em hard hitting drivetrain has put 1885 above the rest so far this season. The word in the kitchen is they’re cooking up improvements to an already dominant machine.
384 (77) - Sporting the fastest human player gear load in the district, SPARKY has returned to their winning ways this season, winning both the Northern and Central Virginia events. A pair of young drivers look to guide this team to victory for years to come. When partnered with another floor-loading gear bot, they proved their versatility in both running the field and also taking care of any rare drops around the airship.
2363 (91) - The Triple Helix autonomous strategy at Hampton Roads was to hang a gear then head to the hopper to grab fuel and attempt to shoot. While the gear step was successful most of the time they were not successful at scoring the fuel. With 3 weeks to fine tune this process a few fuel points could help them seed high this weekend.
836 (89) - Two top 5 rankings, two semi-final appearances, 0 wins. The RoboBees are another example of a CHS team that spent the time to build a shooter that just hasn’t paid off… yet. They’ll look to grab a few fuel points in autonomous and go for 4 rotors as one of the only teams in CHS to do it multiple times.
Others to watch for:
1629 (87) - GaCo has posted their worst average rank in 5 seasons, but they were still top 10 at both events and also made the top three most-valuable teams when ranked by component OPR. A solid autonomous and a gear mechanism which should benefit more than most from a peg upgrade puts this Maryland team in prime spot to disrupt the CHS Platter favorites. Look for them to make their climber more reliable, and also potentially work on firing a shot or two of fuel.
1262 (88) - The STAGs got off to a slow start in Blacksburg but picked up steam in Deep Run, giving the powerhouse #1 alliance a run for their money by taking the Finals to 4 matches. Capable of >1 kPa this green team is not a rookie when it comes to tiebreakers. Left undefended, look for this team to make better strategic decisions while also using their ground pickup to almost solo 3 rotors.
346 (61) - Both Israel and North Carolina championships saw huge upsets by teams who upped their fuel game just enough to get over the 40kpa hurdle. Yet several other high-profile District and Regional events have seen 40-kpa bots fall to a mighty 4-rotor alliance. 346 once showed us they can do 40kPa, but we haven’t seen it since. Was it a fluke? A hoax? A boiler that counted too many fuel? With their performance in Glen Allen putting them just a few kPa away from that magical threshold in several matches, The Platter doesn’t think so. Look for 346 to spoil the rankings and make very strategic picks on Saturday.
Possible Sleepers:
1389 (54) -Seeking to take advantage of the experience in Finals at two district events, The Body Electric will bring a very capable robot to district champs. Rumored improvements to their gear game may be what it takes to move this team from a first-pick into the captain’s role.
4638 (79) - The Jagbots showed a consistent gear cycling robot and superb decision-making on the field. With any above-average gear cycler they could easily hit 4 rotors when left unchecked. If they can improve execution under defense, look for this team to make it deep into the tournament on Saturday.
5587 (89) - This consistent, but slow, gear cycler already has two first seed finishes due to a consistent climb. Their early exits in Quarter Finals at both of their events were a classic “death by serpentine” scenario. If TC Williams can shave a few seconds off of their gear game and improve their scouting, they may just have what it takes to upset the favorites in the tournament.
6334 (91) - These rookies burst into three events this year with very respectable eliminations performances at each. If they can gain just a little more control over their gear feeding under pressure and make their climber just slightly more consistent, the Alumanati will easily go in the first round to a dominant team. A great robot performance coupled with a solid team business plan also (in The Platter’s humble opinion) puts this team on the short-list for Rookie All-Star.
The Spoilers:
These previously-unmentioned teams showed they were capable of specific feats at their latest events. With a higher level of play, expect these capabilities to greatly influence qualification rounds and add some flavor to the tournament.
Solid Autonomous Rotor Routines
401, 422, 540, 620, 1080, 1610, 2199, 2377, 2534, 2537, 2964, 3359, 3793, 4242, 5115, 5279, 5338, 6326
Most-Likely To Hit the 3rd (and possibly 4th) Rotor
422, 619, 620, 623, 1080, 1111, 1123, 1418, 1522, 1610, 1895, 2534, 2537, 3373, 3793, 3941, 4472, 5279, 5945
Have Hit the 4th Rotor in Quals or Elims Already
1111, 1522, 1599, 1731, 5950, 5957, 6584
The Only Other Fuel Bots, So Far
(Statistics show greater than an average of 1 kpa per match at any event)
422, 540, 2377, 3359, 4456, 5338, 5950
Teams Who Climb > 80% of the Time
619, 1418, 1522, 1598, 1599, 1731, 1895, 2199, 2964, 3359, 4242, 4472, 4541, 5338, 5945, 5950, 5957, 6239
